After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 405958 - ZTEST should take extra optional argument
ZTEST should take extra optional argument
Status: RESOLVED FIXED
Product: Gnumeric
Classification: Applications
Component: Analytics
git master
Other All
: Normal major
: ---
Assigned To: Andreas J. Guelzow
Jody Goldberg
Depends on:
Blocks:
 
 
Reported: 2007-02-09 00:36 UTC by Andreas J. Guelzow
Modified: 2008-06-18 01:35 UTC
See Also:
GNOME target: ---
GNOME version: ---


Description Andreas J. Guelzow 2007-02-09 00:36:18 UTC 
given the following values in A1 to A4:  2,3,4,5

ztest(A1:A4,3.5)   correctly evaluates to 0.5
ztest(A1:A4,3.5,1.29) should also be 0.5 but evaluates to 4.93504506253327E-06 which is slightly off (to say the least)
Comment 1 Andreas J. Guelzow 2007-02-09 01:11:31 UTC 
It may not be worth mentioning that to be Excel compatible ztest of course allows a third optional argument that si ignored in the function description, and really returns the right-tailed p-value (rather than the two-tailed p-value). I guess this misdescription is compatible with Excel's documentation that also erroneously seems to claim to calculate the two-tailed p-value.
Comment 2 Morten Welinder 2007-02-09 01:19:01 UTC 
I'm not so sure.

ztest(A1:A4,3.5) is 0.5 because 3.5 is dead-center in the set {2,3,4,5}

1.29 is far to the left of the center of {2,3,4,5,3.5}  (The center is
still 3.5 so ztest(a1:a4,3.5,3.5) is 0.5)
Comment 3 Morten Welinder 2007-02-09 01:26:11 UTC 
Ok, I see.

We should take an extra optional argument to use in place of range_stddev_est
on the range.

Fine, we can do that.
Comment 4 Andreas J. Guelzow 2007-02-09 01:29:36 UTC 
Yes Morten, I had just figured out that my assumption that we were using the same syntax as Excel and OOo was flawed and 3.5 became part of the data set. We should have an optional argument (if we claim Excel compatibility).
Comment 5 Morten Welinder 2008-06-18 01:35:03 UTC 
This problem has been fixed in the development version. The fix will be available in the next major software release. Thank you for your bug report.